Belize has the highest HIV prevalence rate in Central America. National reports state that 28.5% of the individuals infected with HIV are between the ages of 15 and 25. Given the small population size of Belize (222,000) and the large percentage of Belizeans under the age of 24 (64%), the continued increase of HIV cases could have a devastating impact on the adolescent population of Belize. The overall goal of this project is to promote HIV/AIDS risk reduction behaviors among adolescents in Belize by using an innovative approach to the concept of peer education that has not heretofore been utilized in this setting. A peer education program has recently been established in Belize. Forty teens in Belize City went through a four day HIV/AIDS training workshop to become certified as teen peer educators. This program is called PEP/BELIZE (Peer Education Program of Belize). Following monitoring and evaluation, PEP/BELIZE will become a vital part of a comprehensive HIV/AIDS health education and outreach program targeted to adolescents in Belize. There are three specific aims of this study. (1) To monitor the peer education program through the use of both quantitative and qualitative measures. (2) To evaluate the effects of PEP/BELIZE with respect to changes in knowledge, attitudes and sexual practices. (3) To develop a PEP/BELIZE training and reference manual. This manual will serve as a guideline for teens conducting the educational sessions. A theoretical/conceptual model was used to guide the development of the intervention program and will be used to examine the relationship between the dependent and independent variables. This conceptual framework utilizes concepts from The Health Belief Model, Social Learning Theory, and Social Influences Model. An impact evaluation of the peer education program will be conducted by using a quasi-experimental, nonequivalent control group design. To ensure an equal representation of each type of youth group in both the experimental and control groups, the twelve organizations participating in the study will be matched on type of organization, age, gender and educational level. A series of bivariate analyses will be used to test the association between the dependent and independent variables (e.g., chi-square tests, t-tests and correlational analyses). Multivariate analytical techniques, such as MANOVA and regression analysis will be used to test the relationships between multiple dependent and independent variables.